The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources including Apollo company records, LinkedIn profiles, LinkedIn posts, and web search results (Exa, Tavily). This content is authored by third parties and could contain instructions designed to manipulate the agent's behavior during synthesis.
Ingestion points: The skill retrieves external data as described in the workflow of SKILL.md through the sandbase_run_get function after executing research capabilities.
Capability inventory: The skill utilizes sandbase_run to execute data retrieval tasks and sandbase_run_get to fetch results.
Boundary markers: No specific delimiters or safety warnings for the agent to ignore instructions within the retrieved data are present.
Sanitization: There is no evidence of content sanitization or filtering before the agent processes the external data.
[DYNAMIC_EXECUTION]: The skill employs a dynamic discovery workflow where it resolves capability names at runtime via sandbase_discover and then invokes them using sandbase_run based on the discovery results. This involves executing logic from dynamically computed identifiers, though it is restricted to the vendor platform's managed capabilities.

partnership-research